ThinkingCap 减少 Gemma 4 思考令牌
Bottlecap AI 发布了 ThinkingCap-gemma-4-26B-A4B-it，作为 Google Gemma 4 26B-A4B-it 的直接替换模型，旨在减少模型的思考量。在 Bottlecap 自行开展的12项基准测试中，该模型平均减少38.9%的思考令牌，平均准确率下降0.15个百分点，但 AA-LCR 准确率下降了4个百分点。该公司还报告了更低的循环率，并发布了模型的 GGUF 构建版本。
该版本展示了在基本保持 Gemma 4 准确率的同时减少推理令牌消耗的方法，但测试结果来自厂商自报，且 AA-LCR 出现了明显退步。
TL;DR
This is the fifth model in the ThinkingCap series. We reduced how much Google's gemma-4-26B-A4B-it thinks, without compromising the quality of its answers.
- Up to 69.1% reduction in thinking tokens, on average 38.9% measured across twelve benchmarks, for 0.15pp of accuracy
- The first ThinkingCap on a Google model, and the smallest accuracy cost of any ThinkingCap so far
- Looping drops from 3.6% of the original model's responses to 0.8% of ours
- Same sampling settings (thinking on, temperature 1.0, top_p 0.95), same serving stack (vLLM), drop-in replacement
ThinkingCap PRO: want even better efficiency? The PRO version of this model uses up to 87.6% fewer thinking tokens, 62% on average. Faster answers and lower inference costs at scale. Contact us at [email protected].
Our fifth ThinkingCap
We have published four token-efficient models so far: ThinkingCap-Qwen3.6-27B, ThinkingCap-Qwen3.8-27B, ThinkingCap-NVIDIA-Nemotron-3.5-Lightning-30B-A3B and ThinkingCap: Qwen3.8-Flash-Next.
#Bottlecap AI#Google#ThinkingCap#Gemma 4 26B-A4B-it#vLLM#GGUF